System 3: Initial Hypotheses

Michael Anton Dila
5 min readDec 6, 2022


Daniel Kahneman famously described two modes of thought, one “fast” and the other “slow.” The fast mode is instinctual and emotional, linked closely to sense perception and autonomic functions. The slow mode is deliberative, rational and reflective, the kind of thinking we often associate with the phrase, conscious thought. He called the fast and slow modes, System 1 and System 2.

Kahneman’s concern was with the ways that System 1 and 2 can mislead or misdirect us, and his focus was very much on thinking as an individual activity. As phenomena of the brain, individuating these paradigms of thinking is axiomatic. There are, however, kinds of thinking that are not captured by these descriptions, because they are not the phenomena of single minds in insolation, but of thinking beings who are socially engaged and interconnected.

The paradigmatic interaction model of social cognition is conversation. A conversation involves at least two participants who, by engaging each other, sharing a world, values, language, are able to create emergent cognition that is neither limited to nor contained within individual brains, or even in separate minds.

Conversations give humans a mechanism for thinking together. We need a way to describe both how this thinking is different from thinking fast and slow and to describe how people think together through immersive interconnection.

We need a System 3.

Illustration by Dave Gray used with the creator’s permission.

My hypothesis is that System 3 is neither “automatic” as Kahneman describes System 1, nor is it effortful, as is System 2. System 3, to the contrary neither happens by itself, nor do we have to work at it. In fact, System 3 thinking distinguishes itself from the labor of deliberate reasoning in that it feels more like play.

In System 1, we react. In System 2, we engage. In System 3, we begin to think precisely at the point and to the extent that we let go, lose ourselves, and give our self over to something else entirely. Our energy changes, we enter an ecstatic state. System 3 is accessible to individuals, but it is a mode of thinking that by definition involves an engagement with multiplicity.

Bernard Suits characterized games as “the voluntary confrontation of unnecessary obstacles.” Games, I believe, train us for System 3. Whereas games involve us in rule-bound play, there are kinds of conversations and collaboration that invite more improvisational and free play. A conversation can have a purpose, but its content cannot be decided or scripted in advance. In other words, we can plan ahead for how a conversation starts or “what it’s about,” but where it leads cannot be settled in advance. If it were, there would be no point in engaging in the conversation. In this respect, the kind of conversation I am talking about here is game-like: both in the sense that there is no point in playing a game whose outcome is already known (that’s the very point of “playing,” after all) and in the sense that participating in a conversation, like playing a game, means being guided by and observing “rules,” but is something far greater than a series of rule bound exchanges.

System 3 is a mode of thinking that is non-binary and fluid. A further hypothesis is that System 3 is not only vital to engaging with complexity, it is itself irreducibly complex. I am excited to continue to explore, both through theory and practice, how an understanding of and engagement with System 3 thinking is vital to discovery, creativity and to navigating complexity.

One of the fundamental and perennial mistakes we make about our thinking about thinking is to connect the activity of thinking to the existence and behavior of a thinker, singular. There are many reasons for the persistence of this error, from the way we have conceived of minds and consciousness. This is likely also one of the reasons that we have tended to believe that thinking is something that only humans do, and not, for example something that trees and ants may also do.

System 3 may give us an optic that helps us better understand our kinship with other forms of life that proliferate, grow, thrive and organize.

Finally, System 3 is a product of design, in both the general and specific sense. As a general matter, System 3 arises as an effect of a designed system of interactions. More specifically, System 3 thinking emerges as the product of the design of particular forms of interactions, of which I think the largest category is made up of what we call conversations.

I’ll close, for now, with a quick review of my major claims:

  1. There is a kind of thinking we do mainly through immersive interactions with others that is not well described by Kahneman’s concepts of System 1 and 2. I call this social, emergent mode of thinking System 3.
  2. System 3 is both a strategy for dealing with complexity and is itself irreducibly complex.
  3. System 3 is a product of design, it emerges through purposeful, immersive and open-ended interactions which have or acquire intent, but do not aim exclusively at an outcome known or knowable in advance.

This is the first of what I expect will be many more posts in which I further develop and deepen this thinking. I am currently scoping out a project to explore and elaborate a more expansive theory of System 3.

While I am profoundly interested in exploring and articulating the phenomenology (lived experience) of System 3, I also intend a purposefully practical and and participatory project. In this intent, I am in fully agreement with Marx, in his 11th Thesis on Feuerbach: “The philosophers have only interpreted the world, in various ways; the point is to change it.”

A final note about intellectual debts. Let me start by saying that I intend in no way be dismissive of the work of Daniel Kahneman, which I find important and profound. It should be obvious, if not necessarily intuitive, that the very conception and language of System 3 owes him a foundational debt. Furthermore, there are a great many thinkers, writers, artists and others who have made contributions that have made possible my own ability to have any insight at all here. I not only stand on their shoulders, but alongside a great many others with whom I have been in conversation over the last four decades. As I continue this project I will mention and engage with these folks more explicitly and more actively.

Please stay tuned for more “episodes.” Coming soon.

Special thanks to Morgan Plummer, Jen Rice, Michael Moore, Saul Kaplan and Robin Uchida for feedback on earlier drafts.



Michael Anton Dila

Michael is a Design Insurgent and Chief Unhappiness Officer